Free tool · No sign-up
Chinese speech to text
Speak Mandarin and see it written, with pinyin over every word.
What you say shows here.
How does Chinese speech to text work?
Press Speak and talk in Mandarin. Your browser’s speech recognition writes what it hears in Chinese characters, a sentence at a time, and our reader splits each sentence into words and puts the pinyin over them, in tone colours. Tap or hover any word for its meaning. Press Speak again to stop.
Simplified or traditional characters?
Choose 简体 for simplified characters (Mandarin as spoken in mainland China) or 繁體 for traditional (Mandarin as spoken in Taiwan) before you speak.
Can it turn an audio file or a video into text?
Yes, in Immersion: add the recording and press Make a transcript. You get the whole recording written out with pinyin and every word explained, and it reads along as it plays. It needs a free account, since the transcript is kept with the recording. For a YouTube video, our browser extension writes captions for videos that have none.
Which browsers can it listen in?
Chrome, Edge and Safari, on a computer or a phone. Firefox has no speech recognition, so it can’t listen there.
Where does my voice go?
To your browser’s speech service: Chrome sends it to Google and Edge to Microsoft to be recognised; Safari uses Apple’s. We receive only the text it heard, to add the pinyin, and keep nothing.
More free tools
Speech recognition is your browser’s; the words and their pinyin are read by our own reader.